Plant Ecology

, 202:113

Germination behaviour of 14 Mediterranean species in relation to fire factors: smoke and heat

Authors

    • Área de Ecología, Departamento de Biología Celular y Ecología, Facultad de BiologíaUniversidad de Santiago de Compostela
    • Área de Ecología, Departamento de Biología Celular y Ecología, Escuela Politécnica SuperiorUniversidad de Santiago de Compostela
  • L. Trabaud
    • Centre d’Ecologie Fonctionelle et EvolutiveCNRS
Article

DOI: 10.1007/s11258-008-9532-9

Cite this article as:
Reyes, O. & Trabaud, L. Plant Ecol (2009) 202: 113. doi:10.1007/s11258-008-9532-9

Abstract

Fire is an ecological factor that has been present in the ecosystems of the Mediterranean region for thousands of years. Our study was undertaken to acquire knowledge of the effect of fire on the germination of Mediterranean species. We used high temperatures (up to 60°C) and smoke to determine the effect of these factors on the germination of species from the Mediterranean region. The species selected are characteristic of the central Mediterranean basin and are representative of both woody and herbaceous species: Rhamnus alaternus L., Cistus albidus L., Cistus monspeliensis L., Fumana ericoides (Cav.) Gand., Rosmarinus officinalis L., Melica ciliata L., Avena sterilis L., Bituminaria bituminosa (L.) C.H. Stirt., Anthyllis vulneraria L., Coronilla glauca L., Argyrolobium zanonii (Turra) P.W. Balland, Emerus major Mill., Genista scorpius (L.) D.C. and Spartium junceum L. The seeds were collected in Mediterranean shrubland (8) and woodland (6) ecosystems, around Montpellier, France (24°45′N and 3°50′E). Ten treatments were tested: a control, three smoke treatments and six heat treatments. The average germination level (germination percentage) and the average T50 rates (time taken to reach 50% of germination) were calculated. The smoke and heat act in a different way on each of the species. The smoke enhanced the germination of two species, whilst moderate heat increased germination in all of the species excepting R. officinalis, F. ericoides, A. sterilis, A. vulneraria, and G. scorpius. Germination was fastest in M. ciliata and S. junceum and slowest in A. sterilis, E. major and C. albidus. The cues did not significantly affect the rate of germination. Fire modified the germination response of 12 of the 14 species studied.

Keywords

FireMediterranean plant speciesGerminationHigh temperaturesPlant-derived smoke

Introduction

Germination behaviour after fire has been studied in many species from different families in Europe (Trabaud and Casal 1989; González-Rabanal et al. 1994; Reyes and Casal 1995; Reyes et al. 1997; Ferrandis et al. 1998), North America (Keeley 1987, 1994), South Africa (Brown et al. 1993; Brown and Botha 2004) and Australia (Bradstock and Auld 1995; Kenny 2000; Thomas et al. 2003; Tierney 2006). These studies investigated the effect of heat shocks on germination and, in many of the species studied, heat enhanced germination (Ballini 1992; Tárrega et al. 1992; González-Rabanal and Casal 1995; Herránz et al. 1998, 2000). Despite the numerous studies performed, there are many species whose germination response to heat shock is still unknown. Fires are characterised not only by the high temperatures reached, but also by the production of smoke. This is a factor that has been widely studied in Australian species (Dixon et al. 1995; Morris et al. 2000; Read et al. 2000; Thomas et al. 2003; Tierney 2006), South African species (Van der Venter and Esterhuizen 1988; De Lange and Boucher 1990; Brown 1993; Brown and Botha 2004), and North American species (Keeley and Fotheringham 1997, 1998) for at least a decade. However, few studies have been undertaken on the effect of smoke on species from the Mediterranean basin (Casal et al. 2001; Pérez-Fernández and Rodríguez-Echevarría 2003; Crosti et al. 2006); such studies are necessary in order to understand the post-fire vegetation recovery of Mediterranean systems.

We designed a germination experiment with 14 Mediterranean species belonging to five families and four different life forms and have established the following questions:
  1. (1)

    Do the fire-associated cues of high temperatures or smoke affect the germination level? If so, do they increase or decrease the germination level?

     
  2. (2)

    Do the smoke and the heat cues affect the time to germination? If so, by how much?

     

Methods

We selected 14 Mediterranean species: Rhamnus alaternus L., Cistus albidus L., Cistus monspeliensis L., Fumana ericoides (Cav.) Gand., Rosmarinus officinalis L., Melica ciliata L., Avena sterilis L., Bituminaria bituminosa (L.) C.H. Stirt., Anthyllis vulneraria L., Coronilla glauca L., Argyrolobium zanonii (Turra) P.W. Balland, Emerus major Mill., Genista scorpius (L.) D.C., and Spartium junceum L. These species are characteristic and abundant in Mediterranean woods and shrublands. Rhamnus alaternus belongs to the family of Rhamnaceae and is a low phanerophyte. Cistus albidus and C. monspeliensis are also low phanerophytes of the Cistaceae family. Fumana ericoides is a chamaephyte of the Cistaceae family. Rosmarinus officinalis belongs to the Lamiaceae family and is also a low phanerophyte. Melica ciliata and A. sterilis are Poaceae; the former is a hemicriptophyte and the latter a therophyte. The other seven species are Leguminosae, B. bituminosa, and A. vulneraria, which are hemicriptophytes. Coronilla glauca and A. zanonii are chamaephytes, and, finally, E. major, G. scorpius, and S. junceum are low phanerophytes.

The seeds were collected in Mediterranean shrubland and woodland ecosystems, around Montpellier, France (24°45′N and 3°50′E), during their dispersion period and stored in laboratory conditions until the tests were performed (7 months after).

Ten fire treatments were tested: a control, three smoke treatments (Smoke-5 min, Smoke-10 min, and Smoke-15 min) and six heat treatments (80°C-5 min, 80°C-10 min, 110°C-5 min, 110°C-10 min, 150°C-5 min, and 150°C-10 min).

The selection of the smoke treatments was based on the data from experimental burnings (Díaz-Fierros et al. 1990; Vega et al. 2000) and the studies made by other authors (Keeley and Fotheringham 1997, 1998). The smoke treatment was performed with the aid of a Fume 2000 smoke applicator (Casal et al. 2001) based on the methodology proposed by De Lange and Boucher (1990). The apparatus is basically a smoke generator, a refrigeration tube, and a 2.5-m3 chamber that acts as a receptacle for the smoke. Branches of Quercus ilex were used for producing the smoke because it is the dominant species in the Mediterranean woodland and shrubland ecosystems. The seeds of the 14 species were introduced on Petri dishes in the chamber after this had been saturated with smoke and were kept under these conditions for a period of 5, 10, or 15 min.

The temperatures selected correspond to those registered at different depths during natural and experimental fires in Mediterranean environments by De Bano et al. (1977), Trabaud (1979) and Vallete et al. (1994). A forced-air oven was used to apply the heat shocks. In order to guarantee the independence of the replicate, one of each treatment and each species were handled separately. In order not to have a large number of handlings, the 14 species were treated at the same time, that is, 50 different handlings were done (10 × 15). So, five replicates of 25 seeds by each treatment were made. This approach creates blocks of groups of seeds, each of which can function as a replicate for each species (Morrison and Morris 2000). Petri dishes, 9 cm in diameter, were used for incubation. Two porous cellulose filter papers were placed on each dish as substrate. Once the seeds were sown, they were saturated with water during the whole of the 80 days of the experiment. The criterion for germination was the emergence of one 2-mm radicle. Radicle emergence was recorded every 2 days. A controlled environment germination chamber with a photoperiod of 16 h of light at 24°C and 8 h of darkness at 16°C was used. This photoperiod and temperature range make up one of the combinations most similar to the conditions that the seeds of these species are exposed to during the germination period.

The average germination level (germination percentage) and the average T50 rates (time taken to reach 50% of germination) were calculated.

Statistical treatment

The following transformation was used for the percentage data: SQRT[ACOS (no. of germinations/25)]. The factors species and treatment were analysed using two ANOVAs, one for germination percentage data and another for the T50 data. Significant interactions between these two factors were observed in both ANOVAs. When there are significant interactions between two fixed factors, the main effects cannot be interpreted (Quinn and Keough 2002; Underwood 1997). One-way ANOVAs were performed to analyse the effect of the “treatment” factor in each species separately. In those analyses in which significant differences between the treatments were shown, the corresponding Duncan tests were performed to determine which treatments caused the differences detected. For two species (A. sterilis and E. major), post hoc T50 tests were not been undertaken since at least one group has less than two cases.

The level of significance (0.05) was adjusted following Benjamini and Yekutieli (2001), and the level of significance resulting was 0.1625; the program used was SPSS 15.0 for Windows.

Results

Germination level

The control germination levels for the 14 species are very different from one another. There are species with low or very low germination, such as R. alaternus, R. officinalis, F. ericoides, A. vulneraria, C. glauca, A. zanonii, and E. major (0.0–8.0%); others with intermediate germination levels, such as C. monspeliensis, A. sterilis, B. bituminosa, and G. scorpius (37.6, 32.0, 39.2, and 28.0%, respectively); and finally a third group with high germination levels, made up of C. albidus, M. ciliata, and S. junceum (57.6, 78.8, and 72.0%, respectively, Fig. 1a, b).
https://static-content.springer.com/image/art%3A10.1007%2Fs11258-008-9532-9/MediaObjects/11258_2008_9532_Fig1_HTML.gif
Fig. 1

Mean percentage germination (±SD) reached by each species and treatments

Smoke and/or heat affected the control germination level, in some cases increasing it and in others decreasing it. Both factors can enhance, decrease, or leave unmodified the germination of the same species, depending on the doses of the factor received. When the temperatures are not higher and/or the exposure time is not long enough, heat enhances the germination of many of the species studied. However, due to higher temperatures and/or prolonged time, it is common that the germination is inhibited. The effect of smoke is less noticeable. Smoke enhanced the germination of 2 of the 14 species studied (M. ciliata and S. junceum), and moderate heat shocks enhanced 9 species. In contrast, smoke decreased the germination level in C. albidus, F. ericoides, B. bituminosa, and E. major. Intensive heat shocks in C. albidus, C. monspeliensis, F. ericoides, M. ciliata, A. sterilis, B. bituminosa, A. zanonii, E. major, G. scorpius, and S. junceum also decreased the germination level (Fig. 1a, b).

The ANOVA applied to the germination data detected significant differences between the species and between the treatments, and the interaction between these two factors was also significant (Table 1). For this reason, one-way ANOVAs were performed for each species separately, obtaining significant differences between the treatments in all the species except R. alaternus, R. officinalis, and A. vulneraria (Table 2).
Table 1

Two-way ANOVA of effects of heat shock and smoke on germination level and on T50

Source

Germination level

T50

D.f.

F-ratio

P-value

D.f.

F-ratio

P-value

Species

13

160.642

0.000

13

29.717

0.000

Treatments

9

51.934

0.000

9

1.656

0.099

Species * treatments

117

10.615

0.000

76

1.964

0.000

Table 2

One-way ANOVAs for each species of effects of heat shock and smoke on germination level and on T50

Source

Germination level

T50

D.f.

F-ratio

P-value

D.f.

F-ratio

P-value

R. alaternus

9

2.325

0.033

4

0.595

0.693

C. albidus

9

17.298

0.000

6

3.270

0.015

C. monspeliensis

9

37.247

0.000

7

2.808

0.023

F. ericoides

9

5.927

0.000

5

0.962

0.473

R. officinalis

9

0.805

0.614

5

0.606

0.709

M. ciliata

9

12.178

0.000

8

3.476

0.005

A. sterilis

9

8.998

0.000

8

18.983

0.000

B. bituminosa

9

40.883

0.000

6

0.634

0.702

A. vulneraria

9

1.243

0.297

6

0.317

0.916

C. glauca

9

4.849

0.000

6

1.539

0.208

A. zanonii

9

19.833

0.000

7

1.295

0.292

E. major

9

6.168

0.000

5

6.453

0.008

G. scorpius

9

11.824

0.000

6

1.247

0.313

S. jumceum

9

9.783

0.000

6

4.185

0.004

Analysing the role of smoke in each species (Fig. 1b, Table 2), we found that smoke treatments significantly inhibited B. bituminosa germination, and it is noticeable that the longer its exposure to smoke, the greater the degree of inhibition. Heat can either enhance or inhibit its germination. When its seeds are subjected to treatments of 80°C-10 min and 110°C-5 min, this species reaches germination levels of close to 80%. On the other hand, the heat treatments within the range 110–150°C-10 min totally inhibit germination. Anthyllis vulneraria was not affected by heat and smoke treatments and did not show significant differences between the treatments (Table 2). The control germination level of C. glauca (7.2%) is different from the 80°C-10 min treatment that slightly enhances its germination. The intense treatments (110°C-10 min, 150°C-5 min, and 150°C-10 min) completely inhibit germination, although the tests do not detect significant differences regarding the control. The other treatments do not affect germination. Argyrolobium zanonii shows a low level of control germination (5.6%). The moderate heat treatments increase germination a little more, the most stimulating being 110°C-5 min, reaching 32.8%. The intense heat treatments (110°C-10 min, 150°C-5 min, and 150°C-10 min) totally inhibit germination. Emerus major, which naturally reaches a level of 4%, only responds with a significant increase when a treatment of 110°C-5 min is applied, reaching a level of 8%. The Smoke-5 min treatment does not significantly affect the value of the control, and the other heat and smoke treatments reduce germination rates to almost zero (Fig. 1b). On the other hand, the only treatments that produce a significant effect on G. scorpius are the more intense heat treatments, and these inhibit germination. In S. junceum, the moderate heat treatments (80°C-5 min, 80°C-10 min, and 110°C-5 min) enhance germination, reaching a level of 80% with the treatment of 110°C-5 min, and the intense heat treatments (110°C-10 min, 150°C-5 min, and 150°C-10 min) completely inhibit germination. The Smoke-10 min and Smoke-15 min treatments give rise to germination levels that are similar to germination without heat and smoke, but Smoke-5 min significantly increases germination (Table 2, Fig. 1b). In C. albidus, heat and smoke act in two very different ways. The control germination level is slightly increased by a moderate heat treatment (110°C-5 min) and reduced by the severe heat shocks (110°C-10 min, 150°C-5 min, and 150°C-10 min) and by the moderate smoke treatments (Smoke-5 min, Smoke-10 min). On the other hand, C. monspeliensis is not sensitive to smoke treatments and responds to heat shocks in a similar way as C. albidus, i.e., the moderate heat shocks (80°C-5 min, 80°C-10 min, and 110°C-5 min) enhance germination, and the severe shocks (110°C-10 min, 150°C-5 min, 150°C-10 min) inhibit it. Fumana ericoides, although also a member of the Cistaceae family, behaves differently, since almost all the heat and smoke treatments reduce the control germination; only the treatments of Smoke-10 min and 80°C-5 min do not modify it. In R. officinalis, neither the smoke nor the heat treatments modify the control germination values. In M. ciliata, smoke enhances its germination. In general, M. ciliata shows a large increase of germination with the smoke and heat treatments (Table 2), but the 150°C treatments strongly reduce its germination. The germination of seeds exposed to 150°C-5 min decreases by 17%, compared to the control seeds that have a germination level of 57.6%. So, final germination of heated seeds is 40.8% (Fig. 1a). The treatment of 150°C-10 min inhibits germination completely. Finally, A. sterilis registers important increases of germination with the treatments of Smoke-5 min (21% more than the control) and 80°C-5 min (16.8% more than the control), but these increases are not statistically significant, while the more severe heat treatments (150°C-5 min and 150°C-10 min) inhibit it completely (Fig. 1a).

T50

The T50 gives an idea of the time at which germination occurs (Fig. 2a, b). Without heat and/or smoke, the species that germinate sooner are B. bituminosa (reaches the T50 6 days from sowing) and A. sterilis (8 days). Other species with a small T50 in the treatment control are M. ciliata (10 days), A. vulneraria (10 days), A. zanonii (11 days), and E. major (12 days). The first four are grass species, and the two last ones are creeping shrubs. The other eight species studied reach T50 values in the treatment control and vary from 18 to 38 days; generally, these are relatively high trees or shrubs.
https://static-content.springer.com/image/art%3A10.1007%2Fs11258-008-9532-9/MediaObjects/11258_2008_9532_Fig2_HTML.gif
Fig. 2

T50 (in days) and SD reached by each species and treatments

The ANOVA applied to the data of T50 detected highly significant differences between the species, but not between the treatments (species: P < 0.001, treatments: P = 0.099), and the interaction species × treatments was also highly significant (Table 1). The one-way ANOVAs applied to the data of each species detected highly significant differences between treatments in C. albidus, M. ciliata, A. sterilis, E. major, and S. junceum (Table 2).

In C. albidus, the Duncan test detected highly significant differences (Table 2) among the control and 80°C-5 min, 80°C-10 min, and 110°C-5 min treatments, which reduced the T50 of this species from approximately 22 to 14 days. The 110°C-10 min treatment reduced the T50 of M. ciliata from 10 to 6 days, and in S. junceum the 3-Smoke treatments, 80°C-10 min, and 110°C-5 min, significantly decreased the T50 from 19–26 to 32 days, which is the time the control takes to reach its T50 (Table 2). Avena sterilis and E. major did not experience any significant reduction of the T50 reached by the control (8 and 12 days, respectively) with any smoke or heat treatment, but a significant increase of the T50 occurred with severe thermal treatments, A. sterilis regarding the 150°C-5 min one and E. major regarding the 110°C-5 min one (20 and 32 days, respectively).

Discussion

Germination of these 14 species in the absence of smoke and heat cues is extremely heterogeneous, varying between 0.0% in R. officinalis and 72.0% in C. albidus. The fire cues tested were responsible for significant modifications of these germination levels in 12 of the species studied. The two species that were not affected by treatments of smoke and heat are A. vulneraria and R. officinalis.

Smoke noticeably enhances germination of two species, M. ciliata and S. junceum. Similar results were found for a grass from southeast Australia (Poa labirlladieri, Read and Bellairs 1999) and from the Mediterranean basin (Dactylis glomerata, Pérez-Fernández and Rodríguez-Echevarría 2003). Smoke can also inhibit germination. Thus, Read et al. (2000) provide data of inhibition by smoke in Solanum nigrum. These authors conclude that the smoke has no effect on shrub species from a native forest community of New South Wales, many of which had hard seed coats, as occurs in Cistaceae and in most of the Leguminosae species of this study. Other authors (Crosti et al. 2006) showed that 1 h exposition to smoke considerably increases the germination rate, although the germination level did not differ from the control. In this study, the two species whose results were enhanced by smoke enhanced their germination level with only 5 min of exposure, and in S. junceum, smoke also reduced its average germination speed. Keeley and Fotheringham (1998) considered that this negative effect was due to the combined action of the smoke and the increase in the acidity of the environment that was registered over an increased exposure time.

Heat, in moderate doses, both enhances germination in C. monspeliensis, M. ciliata, B. bituminosa, C. glauca, A. zanonii, E. major, and S. junceum and inhibits germination in all of these species at high levels (150°C). Both grasses germinate after 110°C-10 min, and Melica ciliata seed germination is even enhanced by this treatment. Our results agree with those obtained by other authors who worked with different species distributed in both Mediterranean and Atlantic environments. Thus, Trabaud and Casal (1989) and Salvador and Lloret (1995), working with R. officinalis, found that heat did not stimulate the germination. On the contrary, Trabaud and Oustric (1989), Thanos et al. (1992), Valbuena et al. (1992), González-Rabanal and Casal (1995), and Herránz et al. (1998, 2000), studying different species of the Leguminosae and Cistaceae family, found that moderate heat shock stimulated the germination.

The results obtained by different authors in studies carried out on species of the Rhamnaceae family vary according to the species studied. The results of Gashaw and Michelsen (2002), studying Zizyphus mauritanius, are consistent with ours in that they showed no enhancement by heat. On the contrary, Ward et al. (1997), Hanley and Lamont (2000), and Turner et al. (2005) found that heat enhances germination in species of the Rhamnaceae family. No significant differences between the seeds from the control and those treated with heat were found when studying three Atlantic species of the Poaceae family (González-Rabanal et al. 1994). Gashaw and Michelsen (2002), studying two grasses from the Ethiopian savannah, found that germination was enhanced by heat, but found inhibition with severe heat treatments. Our results indicate that one Mediterranean grass is enhanced by moderate heat, and their seeds are also able to germinate after severe heat shock.

Germination in many species is inhibited when temperatures are very high and/or the exposure time is prolonged (Trabaud and Casal 1989; Salvador and Lloret 1995; Herránz et al. 1998, 2000; Reyes and Boedo 2001). It is probable that the same mechanism that produces stimulation under certain combinations of time and temperature could also be indirectly responsible for the deterioration of the embryo and its death above certain temperature thresholds. The high temperatures crack the hard seed case of the Leguminosae and Cistaceae and the covers of the Poaceae seeds, allowing water to filter into the embryo, thus activating germination. If the high temperatures are applied for too long, the heat affects the embryo, drying it to lethal levels.

Some authors have studied the combined smoke-heat effect on germination and have found a diversity of germination responses to fire-related cues. Kenny (2000), studying three Sidney species, found a positive and sometimes synergistic interaction between smoke and heat. Tierney (2006) found a negative interaction between smoke and heat with relation to the germination level of Prostanthera askania. And Thomas et al. (2003) found neutral and positive relationships (some ones additives and other ones synergistic) in nine species of the Sydney region. If the simultaneous effect of smoke and heat had been studied, possibly some of the 14 species would also show some significant interaction with regard to the fire associated cues.

After synthesizing these results, the responses shown by the species to treatment with heat and smoke can be classified into three groups: The first group includes the species that have a notable and significant increase in germination after smoke and heat treatments (M. ciliata and S. junceum). The T50 values of M. ciliata and S. junceum are low. In the second group the species that are significantly enhanced by heat only include: C. albidus, C. monspeliensis, B. bituminosa, C. glauca, A. zanonii, and E. major. In general, the T50 values registered in these species are medium. The third group would be formed by the species that are either not enhanced by fire-associated cues or very little (R. alaternus, F. ericoides, R. officinalis, A. sterilis, A. vulneraria, and G. scorpious), and their T50s are very variable, between 13 and 46 days.

After analysing the response to fire of these 14 species, we can conclude the associate germination cues of heat and smoke affect the germinative response of 12 species. The sign (stimulation or inhibition) and the magnitude of the modification varies among species. At the same time, the effect of smoke and/or heat is usually reflected in the germination level, but not in the germination rate. The modification of germination behaviour is greatest in the species that normally have an intermediate or high natural germination level: C. albidus, C. monspeliensis, M. ciliata, A. sterilis, B. bituminosa, G. scorpius, and S. junceum.

Globally, plant species have opportunities for colonising heterogeneous environments that have been affected in different ways by fire so both those that are enhanced by fire-associated cues and those that are not have distinct ecological niches.

Acknowledgments

The authors would like to thank Dr. Mercedes Casal for the advice given on the design of the experiments and Dr. Margarita Basanta for the suggestions made during the writing of this paper.

Copyright information

© Springer Science+Business Media B.V. 2008